The Unknown Future: To Predict or Not to Predict

[firewalker 10]

  bootstrap said:

The shorter the time frame the more random the markets appear, and the longer the time frame the randomness dies down and only unpredictability remains.

 

  jasont said:

I personally am an intra day trader and can see the patterns of the market change much easier in an intra day time frame better than a daily time frame. I have watched both and continue to watch both daily and intra daily. Is it a case of we as individuals see different things at different time frames?

 

Interesting remarks guys... especially since in my experience I've found the shorter time frame to react more "predictable" than the longer one. Some of the patterns or moves that you find in short term S/R levels, quite easily disappear in the big picture. On the whole, I still find it very fascinating to see how the pieces of the puzzle often fit very nicely together...

[firewalker 10]

  Paul71 said:

People are up in arms around the world over high fuel prices...they have to pay for thier investments...they think their pension is just money that pops up out of thin air...the dumb money pays for itself the pros just move the money according to what the dumb money wants or maybe doesn't want in most cases.

 

Professional intent must surely be the only thing to pick up on in order to predict.

 

I may be wrong, just some thoughts.

 

I think it's too easy to label professional money automatically as the smart money. Who are the professional money? Does it matter? Do they profit all of the time? I don't think so... following their footsteps is obviously the path to take, but it's not because they have bigger 'firepower' they can influence the future path of price on the big timeframe. Given the number of market participants and the amount of money that exchanges hands each day, I think manipulation is much over-rated, but definitely occurs on the shorter time frame.

 

I found this interesting enough to quote here (typos removed):

 

  BlowFish said:

Commitment of Traders Report seems to suggest that the idea retail traders are less 'smart' is not the case. (If you consider smart as being on the right side of the market). The proportion of winners and losers amongst hedgers large players and the small retail traders is more or less the same.

 

  Paul71 said:

It's interesting about how it's all really fuelled. Think about one single person and how much they actually pay into the system, the markets, by way of pensions, mortgages, savings, borrowing etc. Then thier use of commodities, where they work, what thier job is for any particular company, thier earnings for what they produce and so on.

 

I suspect the real dumb money doesn't even know that they are dumb money and have no direct interest in the markets such as being an active speculator.

 

I also suspect that kind of dumb money doesn't mind being called 'dumb' as they aren't the least bit interested But there's nothing wrong with that.

Edited June 18, 2008 by firewalker

[firewalker 10]

Anyone who hasn't voted, please do

 

It's very interesting to see that from the initial 66/33 the results are turning out to become 50/50...

[bootstrap 47]

  jasont said:

Bootstrap, do you mind me asking if you trade daily time frames or intra day? The reason I ask is that I find it interesting to see if those who trade daily believe the market is more random shorter term and vice versa for those who trade intra day.

 

I consider myself a long term intra-day trader. I take postions that are planned to go flat by EOD, but if the action is right the trade moves to the next level.

 

IMO randomness is seen based on the method used and not the time frame traded.

[Eiger 10]

  bootstrap said:

... The shorter the time frame the more random the markets appear, and the longer the time frame the randomness dies down and only unpredictability remains ...

 

I view the markets as chaotic and fractile: chaotic meaning that large movements can start with small influences that may not be easily recognized, amd fractile meaning pattern formations are similar across time frames. Neither time frame is more random than the other.

 

The smaller the time frame, the more susceptible it is to shock. If an unexpected news item occurs, a big move can occur on the small time frame and appear totally random because its size is out of proportion to other moves. On the higher time frames, price may move with rapidity, but the size of the move is more consistent with other movements. Thus, it appears "less random."

 

If the above be true, it begs the question: Is the shorter time frame less predictable than the higher time frame? Every expert always says to start trading on higher time frames first. Is it due to a difference in predictability?

 

Eiger

[firewalker 10]

  Eiger said:

If the above be true, it begs the question: Is the shorter time frame less predictable than the higher time frame? Every expert always says to start trading on higher time frames first. Is it due to a difference in predictability?

 

Eiger

 

I'm not sure that "every" expert says so, but I agree most do advocate this kind of strategy... personally I've found it much harder to determine swing trade positions than intraday ones. But that might be a result of not having studied the higher time frame for the same amount of time as intraday positions.

[bootstrap 47]

  Eiger said:

Neither time frame is more random than the other.......

 

If the above be true, it begs the question: Is the shorter time frame less predictable than the higher time frame? Every expert always says to start trading on higher time frames first. Is it due to a difference in predictability?

 

Eiger

 

To shed a little more light on my perception and where I am coming from. The key to understanding this is the "perception" of randomness. And while this is fun to debate, it really does not matter one way or the other as long as you can trade it and make a consistent profit. Before I get carried away, back to my perception of the randomness of the markets.

 

Compress your timeframe down to the single tick. You can not look backwards, you can only look at the current/last tick and have no knowledge of the past. If XYZ has traded at 102.50 where is it headed? Up, Down, how far. As you move forward tick by tick losing your knowledge of the previous move, you will see the markets as nothing more than a random move each time.

 

Now move up a time frame to a mere two ticks. You now have more information to form an opinion on where the market is headed. The further back in time you move, the more information you have to base a decision on and patterns begin to emerge.

 

but then again, this is just my perception.

[jasont 10]

  Quote

If the above be true, it begs the question: Is the shorter time frame less predictable than the higher time frame? Every expert always says to start trading on higher time frames first. Is it due to a difference in predictability?

 

Could the reason not be based upon predictability at all and more so the failure to manage ones emotions on the shorter time frame. I have found the less time to think about a decision generally makes the decision less logic based and more emotion based to new traders.

[Kiwi 16]

Yes. More time = less impulsiveness; more time to correct mistakes; more time to plan; getting used to being patient.

[BlowFish 10]

  bootstrap said:

I consider myself a long term intra-day trader. I take postions that are planned to go flat by EOD, but if the action is right the trade moves to the next level.

 

IMO randomness is seen based on the method used and not the time frame traded.

 

I wonder how long it would be before someone mentioned the 'R' word personally I don't think markets are completely random imo they tend to cycle from streaky (trend) to more random distributions (back and forth). Just because they are not completely random does not make them predictable of course, they just have tendencies (imo).

 

What was that website you could go to and it would display charts and you had to guess which where real price charts and which where random data? For the life of me I can't find it but it was a fun exercise.

[firewalker 10]

  bootstrap said:

To shed a little more light on my perception and where I am coming from. The key to understanding this is the "perception" of randomness. And while this is fun to debate, it really does not matter one way or the other as long as you can trade it and make a consistent profit. Before I get carried away, back to my perception of the randomness of the markets.

 

Compress your timeframe down to the single tick. You can not look backwards, you can only look at the current/last tick and have no knowledge of the past. If XYZ has traded at 102.50 where is it headed? Up, Down, how far. As you move forward tick by tick losing your knowledge of the previous move, you will see the markets as nothing more than a random move each time.

.

 

It's not because it looks or is random, that it cannot be predicted.

Most will accept Pi as a random number, but that doesn't mean we can't tell what the next digit is going to be...

[jasont 10]

Further on the belief of a random market, I think for a market to actually be random, it would have to be made up of random decision. I personally have no idea of 1% of reasons why someone would buy or sell where they did. However I will guarantee that most traders have a reason to do so. There choice to buy at X price is not simply like throwing a dart at a board whilst blindfolded. They have reason, that alone in my opinion removes randomness out of the market. Unpredictable yes, random I believe not.

[BlowFish 10]

"If we want avert failure in speculation, we must deal with causes. Everything in existence is based on exact proportion and perfect relation. There is no chance in nature as mathematical principles of the highest order are at the foundation of all things. Fardaday said 'there is nothing in the universe but mathematical points of force'." W.D. Gann in how to make profits trading in commodities.

 

I guess we know which way he would have voted!!

 

Edit: he says exactly the same thing when interviewed in December 1909 by Wyckoff in ticker and investment digest.

Edited June 19, 2008 by BlowFish

[DbPhoenix 42]

  jasont said:

Further on the belief of a random market, I think for a market to actually be random, it would have to be made up of random decision. I personally have no idea of 1% of reasons why someone would buy or sell where they did. However I will guarantee that most traders have a reason to do so. There choice to buy at X price is not simply like throwing a dart at a board whilst blindfolded. They have reason, that alone in my opinion removes randomness out of the market. Unpredictable yes, random I believe not.

 

At last.

[Kiwi 16]

  jasont said:

Further on the belief of a random market, I think for a market to actually be random, it would have to be made up of random decision. I personally have no idea of 1% of reasons why someone would buy or sell where they did. However I will guarantee that most traders have a reason to do so. There choice to buy at X price is not simply like throwing a dart at a board whilst blindfolded. They have reason, that alone in my opinion removes randomness out of the market. Unpredictable yes, random I believe not.

 

I like this too.

 

I know predict is foul language for many but I personally believe that you can't trade without prediction. Its just that the prediction is not absolute; it's statistical.

 

For example:

- I predict that I will make money this week (nearing 100%).

- I predict that I will make money today (70%+ because 4-8 trades are involved).

- I predict that this trade will make money (~70% because thats the ratio of trades that win with my "system.")

- I predict that if this is the context (my criteria for "it's trending" and "it's retraced into the right s&r) and this happens (my tested price action) then there is a 70% chance that price will reach target zone before it retraces to where my stop will be sitting.

 

I don't "know" what the market will do; in fact I plan for it doing the opposite of what I predict and what I will do next if that happens. But I do predict with statistical certainty

 

 

Note: For anyone who's ever had a problem with reentering after a loss (often called revenge trading :crap:) I think that I finally eliminated it by planning the trade to follow a loss. Once you have a plan then doing the wrong thing is much less likely.

[BlowFish 10]

Great post Kiwi, though the question was are the markets predictable. Your answers are to a more important question namely "Are my trading results predictable"? This naturally leads to the question "Will I act predictably and according to my plan come what may?". Much better questions for traders to contemplate if they can not immediately answer yes to both.

 

Sometimes it is good to sidestep a question when there are more important considerations at hand however I hoped you did not tick 'yes' based on your answer here.

 

Edit: Work on the one thing that can be made predictable in the markets. You.

[darthtrader2.0 10]

  jasont said:

Further on the belief of a random market, I think for a market to actually be random, it would have to be made up of random decision. I personally have no idea of 1% of reasons why someone would buy or sell where they did. However I will guarantee that most traders have a reason to do so. There choice to buy at X price is not simply like throwing a dart at a board whilst blindfolded. They have reason, that alone in my opinion removes randomness out of the market. Unpredictable yes, random I believe not.

 

The first sentence on wiki as far as randomness to me sums things up perfectly.

http://en.wikipedia.org/wiki/Randomness

"Randomness is a lack of order, purpose, cause, or predictability in non-scientific parlance. A random process is a repeating process whose outcomes follow no describable deterministic pattern, but follow a probability distribution."

I think the difference between randomness in the general way the term is used vs a randomn process causes alot of confusion when ideas drip down to the retail level from the academic world.

To me its hard to come up with a better definition of the markets than

"The markets are a repeating process whose outcomes follow no describable deterministic pattern, but follow a probability distribution."

[Kiwi 16]

Thank you blowfish,

 

The one at the bottom answers the question "are the market's predictable." I just wanted to get there slowly so that people would better understand my argument.

[atto 20]

I haven't read this whole thread yet, but wanted to chime in after what I did see.

 

While "prediction" is a no-no for many traders, if used in the same way as "statistical expectation", I think it's perfectly valid. Case in point: When I enter a trade, I have no idea what direction the markets will go. However, I have a statistical edge in a certain direction (or risk/reward setup), so it is possible for me to exploit that for financial gain. Therefore, I am predicting that the market goes in my direction.

 

If the markets were random, then it would be impossible to profit from the markets exploiting a "bias" over the long term. Likewise, a trader could not make consistant profits using random data (simulating a market) using their bias over the long term.

[MC 10]

  BlowFish said:

Work on the one thing that can be made predictable in the markets. You.

 

The only way to win longer term is by being consistent on YOUR execution of a legitimate edge. I'm still a rookie (less than 2 years) but that's the conclusion I've come to.

[smwinc 20]

As a whole, the behaviour of groups in specific situations can be predicted. This is an accepted theory, and fairly obvious.

 

As long as you can correctly identify the situation, you can put the odds in your favor of predicting the outcome.

 

The problem becomes individuals. The action of a unknown individual can NOT be predicted.

 

A "black swan" event is often when the action of one individual exceeds the affect of the masses.

 

This can include government intervention, the death of someone important (like a CEO), a fat finger trade, etc.

 

Knowing this inherent weakness, a trader should attempt to mitigate the risks by trading in markets that are less able to be impacted by one individual.

 

A thin market will always be less predictable than a thick market over the short term.

 

One stock will always be less predicable than a large index.

 

You can also mitigate the risks by certain times that you trade during different markets.

 

E.g. In a thin market, the last 10 minutes is often very hectic as brokers force through late orders. This often will be completely random. Or perhaps more accurately, the degree of randomness outweighs the degree of predictability for most trader's risk tolerance.

 

Etc.

[atto 20]

  smwinc said:

A "black swan" event is often when the action of one individual exceeds the affect of the masses.

This is a wonderful point. I recommend anyone who takes any risk (everyone) to read The Black Swan. It's truly insightful and thought provoking concerning risk and how bad we are at managing it.

 

(P.S. I absolutely love it when people use the alternate definition of "affect" correctly, assuming you meant to .)

[BlowFish 10]

  Kiwi said:

Thank you blowfish,

 

The one at the bottom answers the question "are the market's predictable." I just wanted to get there slowly so that people would better understand my argument.

 

Oops sorry, I am always the one that gets all excited when the aha moment comes and then spoil the carefully constructed presentation by blurting out the 'punch line'.

[MC 10]

I can't lie...it's cracking me up that NOBODY has chosen number 3. :\

[firewalker 10]

  MC said:

I can't lie...it's cracking me up that NOBODY has chosen number 3. :\

 

To be honest, I kind of expected a fight between 1 and 2

looks like it's turning more into 60-40 again

 

Which is interesting imo, because most of the members who replied and posted seem to advocate option 2.